Frederic Branczyk
Frederic Branczyk
Actually, one more thing, could you share a memory profile? Then we could see if my suspicion is true in terms of where the memory is spent.
Did you mean the source label of the first relabel config to be `__prefix__`?
Makes sense, so you meant `__path__` if that existed?
> Is the following example a Prometheus configuration file? Because while static_configs has been [defined in Parca](https://github.com/parca-dev/parca/blob/39b958f7fee944d38d7f398a57cdc106c139eaeb/pkg/config/config.go#L168), according to the [main parca.yaml file](https://github.com/parca-dev/parca/blob/39b958f7fee944d38d7f398a57cdc106c139eaeb/parca.yaml#L14) and [tests](https://github.com/parca-dev/parca/blob/39b958f7fee944d38d7f398a57cdc106c139eaeb/pkg/config/config_test.go#L44) it is supposed to be...
Is there something specific about Java Flight Recorder that you're looking for or do you think this can be generalized as Java support?
Thank you for clarifying! I think java support shouldn't be too hard, especially since it's JITed code. It's not as easy as I would like it eventually be, but https://github.com/parca-dev/parca-agent/pull/101...
We should do that regardless!
Agreed that pprof labels are the place for this. Maybe a long shot but we're already walking the stack anyway in eBPF, so I imagine with some languages (like Go)...
Yeah, I think it should be possible to do it both ways.
Briefly discussed on discord: this is already reflected in the URL but not the state within these, eg. what was searched and what was the selection within an icicle graph.